September 2, 2004
Hubert H. Humphrey Building
Room 705A
200 Independence Avenue, S.W.
Washington, D.C. 20201
Proceedings By:
CASET Associates, Ltd.
10201 Lee Highway, Suite 160
Fairfax, Virginia 22030
(703) 352-0091
TABLE OF CONTENTS
Report on Executive Subcommittee Retreat
Race, Ethnicity and Socioeconomic Position
Reports from Subcommittees and Work Groups
DR. LUMPKIN: We're going to try to get started. This is the second day of the National Committee on Vital and Health Statistics meeting. We are going to start off with introductions. My name is John Lumpkin. I am the Chair of the NCVHS, and I am Senior Vice President at the Robert Wood Johnson Foundation.
(Whereupon, introductions were performed.)
DR. LUMPKIN: I think that is everybody. Let me just do a quick review of the agenda for the rest of the day. The opening discussion, I am actually going to merge the NHII update and the highlights of the executive subcommittee retreat to eliminate redundancy, since that was a big part of the executive subcommittee retreat.
Hopefully we can get back on schedule for the populations panel. Then there is an item at 12 p.m., where it says lunch. We are going to make that a short break, and then we are going to work through lunch with the goal of finishing earlier than 2:30.
The executive subcommittee retreat was held again this year in the friendly confines of Princeton, New Jersey. Elvis was not in attendance. For those of you unfamiliar with local lore in Princeton, Elvis is a peacock that has been wandering around Princeton and the grounds of the Robert Wood Johnson Foundation this summer. He has yet to be apprehended by the Animal Control people, but did not make an appearance for the executive subcommittee retreat.
We started off the retreat with some discussion about the environment which we are in. This is really an exciting time, which is in many ways very positive, but it also forms the basis of an ancient Chinese curse: May you live in exciting times. Or is it interesting times? It is both interesting and exciting.
There are a lot of things going on as part of the transformations around the National Health Information Infrastructure, around the adoption and interest in information technology. Many of us, or those of us who have been on the committee for any length of time, remember that when we issued our report in 2001 on the National Health Information Infrastructure, that subsequent to releasing that report, we then proceeded to try to get some attention to that report. We tried to meet with various senior level individuals within the Department of Health and Human Services and others, without any success.
A key meeting occurred in November of 2002, almost a year after the report, when the Institute of Medicine released a report on rapid advances in health care, a report commissioned by the Department of Health and Human Services.
At that workshop that was convened by the Institute of Medicine, the Secretary attended for an entire day, which is unheard of. There was a panel that was held in one of the areas identified in the Institute of Medicine report, which was use of information technology.
At that panel, the importance of the National Health Information Infrastructure and paying attention to that, and transforming the health care system to enhance effectiveness, efficiency and quality, was raised, and was really a turning point in the attention of the Secretary in the Department. Subsequent to that, the Secretary has been significantly engaged in issues related to information technology.
That would have been big enough in and of itself, but anyone who has been paying attention to current Presidential campaigns will note that both candidates have platform positions in relationship to health information technology. There was a recent editorial op-ed piece, I can't remember if it was in the Washington Post or the New York Times, authored by Senator Frist and Senator Clinton, not a group that one would expect them to agree on most anything, again noting the importance of information technology and transforming the health care system.
So all the things that we have worked for, the vision that we have worked for, has now been adopted, which is certainly a good thing, and it is also a challenge for us. It is a good thing for us as a committee obviously, because now as we begin to do our deliberations, the attention that is paid upon our recommendations, which has been very high in the past, remains to be high.
The challenge for us as a committee is that we are no longer the only force pushing for many of these issues. Recently at the National Health Information Infrastructure Conference, David Baylor, who had been appointed in May, played a lead role in bringing forward the Department's strategic plan in relationship to health information technology.
The field is changing very rapidly. The rates of adoption of electronic health records of different flavors have increased, and we are in a period of rapid transition. This has been a problem for many of our subcommittees and work groups. It has been a problem for the NHII work group, because we pick a subject, we talk about what we can do on it, and by the time we get around to it, much of what we had talked about has already been adopted or moved forward by the Department or industry. The same is also becoming true of some of the work of the quality subcommittee, as other issues are being adopted by the industry, and quality is beginning to be placed on the table, with CMS taking specific roles in pay for performance and other related initiatives.
So that is really the background for which the executive subcommittee held their retreat, recognizing that information technology is being widely accepted as a key part of the strategic foundation, both politically and administratively throughout the health care system.
We also had as the environment the Medicare Modernization Act, which specifically cites roles for the National Committee on Vital and Health Statistics, which continues an ongoing process for Congress identifying specific roles for our committee. As you know, if that wasn't enough, the Secretary then took the direction from MMA, particularly in relationship to e-prescribing, and said that is good for Congress to tell us to do it in that time frame, but we are going to do it sooner. Thus, the Herculean task that was before the Subcommittee on Standards and Security in coming forward with today's letter on e-prescribing, and that is just the beginning of some of those activities.
The continued role of the committee was highlighted by the fact that in the recent summit that was held by the Department of Health and Human Services, the health information technology summit which was combined with the National Health Information Infrastructure Conference, selected the National Committee on Vital and Health Statistics to host a third day of that event. The work group on the National Health Information Infrastructure received the reports from the various breakout sessions of that conference, to assure that those reports can get out and be discussed widely. The results of that you saw yesterday with the adoption of the letter to the Secretary, which had as an attachment the reports from that conference.
There are certain observations that were made, summary observations that I want to go through, and then talk about some of the specific discussions at the executive subcommittee retreat.
The first is that the NCVHS is in a unique strategic position to address both the e-technical as well as the population-based aspects within one group. This is very unusual, since most of the organizations in this space at this time tend to have a certain focus. They may be focused on health care, they may be focused on population health. We have the advantage as a committee based upon our historical activities, as well as our current membership, to cross those two areas and look at where they intersect. That is something that is special and unique about us, and something that we need to bear in mind.
The second is that the power of the committee is not only in our vision, and certainly with the National Health Information Infrastructure and other areas of activity, we have had the vision to point out where this field needs to go, but also in the specific recommendations that we make. We have a very high level of adoption of our specific recommendations by the Department of Health and Human Services, and because of that, it is not only the power of our visions, but also our specific recommendations.
Third is that the committee should encourage attention to research and articulate a health information research agenda for the Department. I'll come back to that in just a second.
Finally, along with traditional extensive data system collection, we now require more nimble analytic integration and more analytic intelligence. This means that pretty much the environment is being created. One of our former chairs of this committee, Kerr White, was once noted to say that we have the incredible capability to create large volumes of health data untouched by human minds. Our challenge as a committee is to recognize that not only is it important to collect the data, but it is also important to analyze the data.
The conceptual model -- Harry and I were having a discussion over dinner last night; our goal for the electronic health record and the National Health Information Infrastructure is not to create a collection of health data about an individual, much as you would think an encyclopedia is collected by a collection of articles, but that this book really needs to tell a story. It is not good enough to collect the data, but it also has to help with decision making and making sure that this story leads to high quality, efficient and effective health care.
What did we talk about at the committee? First was the area, maintaining attention to population health as an overarching issue for the full committee. We talked about some activities related to that.
We have had some successes, for instance, partially through our work, but also through some of the initiatives through AHRQ, where we have assisted, that they are now working with health plans, collecting information on race and ethnicity data, and using that as part of a collaborative to improve the quality of care and reduce disparities. We are hoping that we are getting some success in working with NUBC, recognizing the functionality of collecting race and ethnicity data as the business case is beginning to be made as health plans look for this data to try to reduce disparities in health care delivery.
Various subcommittees and work groups will need to continue to consider population issues in their review and deliberations. We have had some important successes that we have moved forward. The quality work group for instance met with the population subcommittee just yesterday to talk about some of the issues and candidate recommendations related to race and ethnicity data, and other aspects of collecting important information such as socioeconomic position. The privacy subcommittee has expressed similar interest in coordinating with the population subcommittee.
I have already talked a little bit about the importance of analysis, and we need to continue to keep that in mind as we assure that we continue to look at not only how we can collect data, but also the analysis. Within the population perspective, this becomes increasingly important that we look at the lessons that are learned, and talk about how we can merge the concepts of the health of the population.
To move the population aspect into health care is going to be a very important component. Clinicians need to be aware of health trends within their community. They need to be aware of risk factors within their community, whether they be environmental such as pollution, but also risk factors associated with the ability to exercise. Current models of enhancing quality improvement such as the chronic disease model developed by Ed Wagner and others, extend beyond the health care facility into self management and into the community. It doesn't do much good for a physician to prescribe exercise to a diabetic who lives in an inner city community when there are no resources for them to engage in physical activity. So the intersections between population health and other activities as a committee remains as something that is important for us to look at.
We also began to have some discussions with the Board of Scientific Counselors, and then based upon that, we have come up with some joint activities that we will have over the next year. In discussing the National Center for Health Statistics Board of Scientific Counselors, we are proposing that we would have a joint hearing on the quality of life measurement in 2004 and 2005, in that period of time, a joint hearing between us and the NCHS Board of Scientific Counselors. We are hoping to schedule a joint meeting of the executive subcommittees of both the Board of Scientific Counselors and the NCVHS in 2005, looking towards a joint meeting of the full committees in 2006. This will continue our organic relationship that we have with the NCHS Board of Scientific Counselors, given the history of the NCVHS in areas that overlap.
We also noted that there is specific interest in our activities by Dr. Braylor and his role as the national coordinator for health information technology, particularly looking at privacy issues, regional health information organizations, infrastructure and making that some of the priority work for the committee in the upcoming year or so.
The executive subcommittee noted the need to have better relations with some agencies outside of the Department of Health and Human Services, particularly the Census Bureau, the Environmental Protection Agency and the Department of Homeland Security. Letters are being prepared asking for key contact persons with those agencies, which will be sent out by the chair of the NCVHS.
We are also agreed to submit and to begin the process of developing a biannual report covering major activities of the committee covering 2003 and 2004.
Finally, in the area of research, one of the things that we felt as a committee that we can contribute to is that there are many areas where we are beginning to realize that there needs to be better research and understanding. Part of our mission in relationship to the Department is to make recommendations on where research ought to be performed by the Department of Health and Human Services. So as part of the process of developing this annual report, we are going to be asking the subcommittees and work groups to identify items for a recommended Department of Health and Human Services research agenda. We are going to have further discussion on this issue in our November full meeting of the NCVHS.
So we had what we thought was a very interesting and exciting meeting of the executive subcommittee, and a lot of interesting discussion. At this point, are there any other members the executive subcommittee who would like to add to my summary? Or do we have questions and discussion based upon the issues raised? Marjorie.
DR. GREENBERG: It really was an excellent meeting, though we decided we missed the peacock. We did appreciate RWJ's hospitality.
Just on the issue of the research agenda and the incorporation of that into the upcoming biannual report, I think we observed -- I just wanted to expand on that a little bit. We observed as John said that a number of the committee's specific recommendations have not fallen on deaf ears. I don't know what the positive side of that is, but they have been well received. They have been adopted, or they have driven policies that have been articulated by the Department.
But we felt that often, our reports do have some recommendations for further research that is needed, but those kind of get lost, and maybe they are not directed enough or we don't have a customer for them, but often the committee recognizes that more research is needed in a certain area, and the committee of course is not able to carry that research out itself. So we thought it might be more effective to try to pull those together and try to set some priorities as a research agenda, and then we could focus on how can we also try to get some uptake of these recommendations.
So I think that is an excellent idea, and I just want to encourage the different subcommittees and work groups, if you are meeting between now and the November meeting, and I know many of you are, or having conference calls or whatever, it is something that you might be thinking about. In particular, maybe the lead staff could pull out recommendation letters that have come out of your subcommittees or work groups in the last maybe few years, if there have been some research recommendations. You could also be thinking about some additional recommendations that you might want to put on the table for our discussion in November, and for inclusion into the annual report, which Susan Canaan will be preparing for us.
So I just wanted to mention that.
DR. HARTMAN: Mr. Chairman, that was a very good report. I appreciate the work that the committee is doing.
You made an implication that our proceedings were perhaps not moving as quickly as other individual groups or large groups' proceedings, that we were flying a turboprop and others were flying jets, so to speak, in the pace of coming out with things and products. Maybe I misunderstood you, but I wondered, if that is the case and we are -- our deliberative process is perhaps slower than other parts of the government, is there something that you all thought we could do about that, that would be helpful in keeping up with the rapid changes that are occurring.
DR. LUMPKIN: Actually, I'm glad you asked that question, because that was not the intent. I think the intent is that we need to be selective in what we choose to do. We need to look where we can contribute.
A good example is the letter we had yesterday on the NHII. There were two areas. One was on conformance testing, and the second one was on metrics, where we chose not to do any action and would monitor in six months. When we raised those, the staff from the national coordinator's office said that is something we are working on. So if they are already working on it, there really is no need for us as a committee to suggest that they do it.
So what we have seen is that there are areas where we can contribute, and we just have to be a little bit more selective, because the field is moving. What we have always done is to pick areas where there seems to be a void, and we just have to be a little more careful in doing that now.
DR. STEUERLE: John, I'm trying to interpret where you are going with your statement, too. I see it as -- maybe I am misinterpreting, this is my first year on the committee, but it is a statement trying to flush out with the executive subcommittee and with the larger committee as to where does this committee as a whole go, and where does it at the margin best direct its resources, given changes, including the successful implementation of a lot of our recommendations in the past, and the development of staff at HHS who are taking this and working on it on a daily basis, which we are not.
One item that seemed to be emphasized in your talk that I haven't noticed a lot is back to this issue of research. I am dodging the question of NHII providing a huge base for research, but outside of that, the only place I have seen a lot of discussion on research is with Vicki's group, and that is with respect to some of these population issues.
Then I heard you say stuff about contacting Census and other agencies, which -- at least if I am hearing you right, is saying there is data that maybe need to be developed there, maybe there is research we need to be thinking about here, we need to sponsor.
All this is to say, when I think of a research agenda that a committee like this can best address, it is what don't we know and why don't we know it. It is not performing the research; we are trying to figure out where at the best margin.
At least at the back of my mind, I can think of a lot of items in my areas, which are somewhat narrow. I can think of agencies you didn't mention that are worth contacting, like social security and matching Medicare and social security records, as one example, that are worthwhile pursuing. I'm sure that some of the other members of this committee all with their background have a huge number of items, not necessarily of research that they want to do, but where possibly marginal changes, possibly even inexpensive changes, can lead to substantial improvement in what we know.
I haven't heard that discussion before today, and I wonder if that is a direction in which you are heading. Does that make sense?
DR. LUMPKIN: Yes, and I am going to call on Jeff, because he looks like he wants to respond to that.
MR. BLAIR: I wasn't going to respond to that, so if there is discussion about that, I'll save my comment for after that discussion is complete.
DR. LUMPKIN: I think that the way you posed it is correct, that we are not looking for an agenda of what research this committee will do, because we are not a research organization. But to make recommendations to the Department of Health and Human Services on where there are areas that will help move forward the agenda in research.
Let me give an example. If they were doing a new system of e-prescribing, maybe they should do a pilot test. Well, MMA already came up with that, so we don't have to make that recommendation, but there are other areas where very likely research ought to be done.
I'm trying to think, there was something that was coming up in populations that there was some discussion, that jumped to my mind and I said, that is an area the Department ought to be doing some something in. We haven't really consciously tried to focus that as a research agenda, even though as Marjorie said, it has come out in various areas of our report. I think what we want to do in the future is to highlight that more.
I don't think that there is necessarily a tie-in with looking at how we reach out. We are looking to other agencies less on how we can enhance data sharing, getting access to databases, as much as we are looking at players who may play a role in setting the health information agenda.
We have already reached that with a key player. We are already working with the Veterans Administration and the Department of Defense through the CHI. As we looked at other agencies that play an important role in directions, certainly homeland security in relationship to population issues, particularly in surveillance of threats and the need to tie that within the public health surveillance that is being done in certain sectors in population.
The Census Bureau, which is for us something we haven't done, and then EPA, because EPA does a lot of measurement of chemicals, and there hasn't been a real good connection of what those chemicals mean in relationship to human biology, as opposed to what those chemicals mean to the environment. They are not necessarily the same.
So those were the three agencies that were selected for potential activity. Then as we develop relationships with them, we will look forward to potentially adding others.
DR. VIGILANTE: Just from a very pragmatic point of view, we are looking for areas of intersection and overlap. If you take a study like the National Children's Study, which is looking to follow 100,000 maternal-child pairs for 20 or 30 years, the largest cohort study that has ever been done, realizing these children are going to see multiple, multiple providers in many different places over their lifetimes, it becomes clear that that kind of study is going to be very hard to do without an NHII of some sort by which these providers can share information with the folks who are coordinating the research.
So I think there are areas in which our NHII agenda and research agenda go down a final common pathway. To the extent that we can help envision that, I think it is an important contribution. The same can be said of the cancer bioinformatics group getting together 50 or 60 national cancer institutes as a research analog of the NHII. Our role there I think is relevant.
MR. BLAIR: John, you indicated that the world has changed and that many of the things that you wish for, be careful what you wish for, you just might get them. Things that this committee has wished for for the last ten years have started to come about in these last few years. The rate of change has increased dramatically.
You properly pointed out when you responded to Richard that we have to choose the things that we decide to explore very carefully, so that we don't hold back the progress that we have fought so hard to try to create in terms of the sense of momentum and sense of urgency. That is now happening, and we don't want to hold that back.
Maybe at the next executive subcommittee meeting, we should consider one other thing. That is, we apparently -- the work we have done has been good enough that Congress has come back to us again and asked us to do things on e-prescribing and probably will come back to us again on maybe some other things. It is like the consequence of success.
The thing about it is that the time frames are different. The e-prescribing activities may not be a unique circumstance. The sense of urgency, the time frames, the resources put tremendous pressure not only on the subcommittee members to move at that rate, but also on the support staff as well. We all were dedicated and committed to it. We became a real team.
I think we were proud of what we were doing. We weren't sure we would be able to get this letter through. But I think there are many of us that question whether this type of intensity can be sustained. However, if we ask that question, the needs of the nation right now seem to be calling on us, or maybe some other entity, to put forth this effort on these issues. Someone has to do it.
So I think the executive subcommittee might investigate what it might take to get ready to support this faster pace of recommendations if the nation calls on us again, where we could provide greater support to the staff, greater support to the subcommittee members, because maybe we are in a new world, where we have to operate a little differently, and we have to examine our ability to meet the needs moving faster.
DR. GREENBERG: Very good point.
DR. LUMPKIN: Good point and well said. Mark.
DR. ROTHSTEIN: I want to raise a broader question that has been influenced by my work on privacy and confidentiality. That is the question of whether the committee as a whole is going to be viewed as an advisor to the Secretary of HHS, or more broadly as an advisor to the federal government in general on issues related to health information. Those are quite different things, of course.
In our subcommittee, for example, we have been really working for the last two to three years certainly almost exclusively on HIPAA related issues, because of the timeliness of the need to develop those recommendations, while letting a variety of other possible topics of inquiry get to the back burner.
The question of moving things forward besides in NHII and e-prescribing and so forth raises the question of how broadly we should be construing our mission. There are many other areas where we could focus, where no one is looking at those issues, but again, I don't want to be on the lookout for work to do, and it is a question of what the expectations are of NCVHS and the resources and ability to do that.
DR. GREENBERG: Jim isn't here. I'll respond -- I don't know about respond, but reflect on what you said. I thought you were going to go even further. I think through its entire history, the committee is first and foremost an advisory committee to the Department, but has had you might say a bully pulpit or a reach that went not only through the whole federal government when it came to health information policy or health statistics policy, which was more of its focus during its earlier years, but outside of the government.
I still get calls from people about the uniform hospital discharge data set, which was developed by this committee, people in hospitals, people in other parts of the private sector. So I think there is no doubt that people well beyond even the federal government have looked at this committee over its entire history.
I think on the other hand, the work has to be responsive obviously first and foremost to the Department, but health information is a huge part of our economy and many other agencies are involved with it, and we have worked with those agencies. Gene mentioned the Social Security Administration. We have worked less with maybe Education or Housing or Transportation, but if you look at functional status information, that overlaps all of those as well.
So I think it just gets back to, we have a very broad charter, but I think it gets back to what John and Jeff were saying, trying to identify where we can make the biggest impact, and then what resources we have to bring to bear. But I think you are probably going to be hard pressed to find a health information issue that doesn't have some relevance to the Department, even if it has relevance to another department or other parts of the health industry.
DR. ROTHSTEIN: I just want to give an example of investigating the intersection between FIRPA and HIPAA. We learned that there is no counterpart committee at the Department of Education that deals with FIRPA issues and privacy. I don't know whether that is useful to the committee's mission to get into something like that, but it is certainly very closely related to what we do in terms of control of information of a medical nature and school records.
DR. LUMPKIN: I think as a committee, our charge and our charter is to advise the Secretary. That hasn't stopped us before. I don't think that is necessarily what is stopping us now.
I think the point that Jeff raised is the operational one. We do have a primary job, and as long as we take care of our primary job, we can have as many hobbies as we want. But within that context, a lot depends upon the time and availability of the individual -- and interest of the individual members of the subcommittee and the staff.
So I think that there is no hard and fast answer. But again, our focus is that we report to the Secretary. So any recommendation that we are going to make is going to have to go to the Secretary, which makes it difficult if we are making recommendations for things that are wholly under the purview of other agencies.
DR. GREENBERG: We have sent letters to the National Association of Insurance Commissioners. But that was really the point I was going to make, too. The recommendations go to the Secretary. I don't think it would probably be appropriate for this committee to start sending recommendations directly to the Department of Education, but to explore these intersections, because it does intersect with HIPAA, et cetera, and then to send a recommendation saying that there needs to be more coordination between the two departments, and then to offer good services to hold additional hearings or something, I think would be very appropriate, if you had that high of a priority.
DR. LUMPKIN: At this point, I am going to -- this is not an end to this discussion; this will be an ongoing discussion, as it has been for as long as I have been on the committee.
DR. GREENBERG: I wonder if Aldona might want to say something from the Board of Scientific Counselors point of view. I realize that you all are meeting within a few weeks.
MS. ROBBINS: Yes, and I know you all are under a tight schedule, so just quickly, we are going to be meeting next week and getting into organizational mode. This will be the start of our second year. We are looking to follow some of the patterns of what I have observed here, in terms of working groups and whatnot.
I think we would very much like to -- as Dr. Lumpkin said about having joint hearings and then looking towards a joint meeting. I think at least what I am going to take back to the board is that because of all of the work that you are doing which we have just heard about, that five or six years ago was probably beyond imagining, but we are probably the ones that need to pick up the pace on looking at the federal health statistics role that this committee used to have the time to do, but no longer. It has been eclipsed by events.
I think that from talking with our chairman, June O'Neill is really the emphasis that we want to put on, is to start looking at the data sets and seeing how can they support this new world that we are moving into.
DR. LUMPKIN: Thank you. I am going to turn it over to Vicki.
DR. MAYS: Thanks, John. We are going to have a panel this morning. This panel evolves from the last time that Populations had a report. People were asking questions about socioeconomic position, about race, so the goals of the panel this morning are to share with you some basic information about some of the concepts that are important in the consideration of population health. We see this as being in support of the full committee's efforts in the area of population health.
The areas that the Subcommittee on Population often brings to the committee around the collection of data on race, ethnicity and socioeconomic position, like Harry's discussion with John, parts of what we think is important is not just that we are looking at these as collecting a field of data. We now have the race question answered. But today's presentation, what I hope it is going to do is also help you to understand that we want to tell a story, and that some of the data that we collect, there is a complexity to it.
Sometimes these recommendations look quite simple, but they actually have a lot of complexity to them. We struggle with this in the Subcommittee on Population, and I thought today it might be helpful for you to get some insight into what these issues mean, and even a little bit about the reasons that we need to collect this in order to better the health of all Americans, not just racial and ethnic minorities, but understanding for example where our health care system doesn't work well, where there are inequities, where there are inequalities. Sometimes when we shine the light on disparities, what it does is help us to have a better system for all.
So with that as the background, today we have two members of the Subcommittee on Population. Virginia Cain is the liaison for us here, so we are quite fortunate to have them both.
Suzanne Heurtin-Roberts is with the National Cancer Institute. One of the things that Suzanne does in her day job -- we talk about having hobbies -- what she does in her day job is, she is the coordinator for health disparities at the National Cancer Institute. Suzanne has a portfolio in which there are now -- how many population centers?
DR. HEURTIN-ROBERTS: Over eight centers. Four plus are funded by NCI. We have cofunders.
DR. MAYS: So population health is something that is definitely on her plate, particularly in terms of the area of health disparities. Virginia Cain you know from sitting here as the liaison from NIH, is the acting associate director from the Office of Behavioral and Social Science Research.
Dr. Heurtin-Roberts is going to talk about the issue of race and ethnicity, and Dr. Cain is going to talk about socioeconomic position. Suzanne, why don't we let you start? Thank you, both of you, for the time and energy that you put into putting this presentation together, because we know you have day jobs, too.
Agenda Item: Race, Ethinicity and Socioeconomic Position - Suzanne Heurtin-Roberts
DR. HEURTIN-ROBERTS: A couple of them. I have titled this for want of a better title race and ethnicity in health and biostatistics, but a lot of the points that I am going to make are just as appropriate for the use of the race and ethnicity concepts in any kind of research or data collection.
Why is there confusion? God knows, there is lots of confusion. One of the reasons is that these are scientific concepts that are used in everyday speech. We may say what we say in our publications, but in everyday life, everybody is going around using these terms meaning a host of different things, and researchers are part of the rest of the population, and we are influenced by that. So that is one of the problems.
Another problem is that these are used interchangeably and often without definition. I'd say probably most frequently without definition in research in the past. They are used atheoretically, in which case there is no explanation for why they are used, why we think it would make a difference. They are just put there to collect the data, but there is really no thought given to what we mean by this. Then as you will see, there are lots of difficulties in measurement. So I can't say that I am going to clear up the confusion, but I will let you know why there is confusion.
For race, I'm sure all of you have heard the discussions about whether race is a biological entity. I would say that race is certainly not a genetic entity, in terms of the categorization of races cannot be determined by any genotype. Race is the categorization of parts of a population based on the physical appearance of people due to particular historical, social and political forces.
The thing to emphasize here is that race is the categorization of a population based on physical appearance. There is a lot of history that goes into this. 19th century anthropology, that we are trying to pay for our sins -- I'm an anthropologist -- a lot of foolish things were done, and destructive.
Race is not genetically defined. There are no racial genotypes to delineate boundaries among races, none. There are none. I'll say it again. There are none. When you look at genetic distribution of the human population, there is no clear boundaries as to where you would delineate one category versus another category; we all blend into another. Race is socially defined, based upon appearance. Our appearances may or may not represent historical breeding populations. I know I sound like an anthropologist saying that, but we don't all get married, but we do share our genetic material.
There is the case of parallel evolution or parallel development of physical characteristics, where there has not been a relationship between two populations. For example, persons in Melanesia, which is a series of islands in the Western Pacific, can often be mistaken for persons from sub-Saharan Africa, but there has been no contact, at least in recent historical times, to warrant this. But people do look similar. They are dark,they have a lot of the same features. So appearances may represent some common gene pool in the past, or they may not. It is dangerous to attribute similar appearances to a common heritage.
The determination of race is subjective. That is another big problem. You can have internal identification, that is, I say, I feel, I am of a particular race. There is what I feel, but then there is what I express to the outside world as my race. Then there is what other people consider me to be, how other people identify me.
All of this is contested. There are no clearly agreed-upon phenotypical criteria to delineate boundaries among groups. Phenotype, I think everybody knows, means the appearance, one's physical appearance, the way one's genes are expressed. There are no clearly agreed-upon phenotypical criteria to delineate these boundaries. As a matter of fact, these boundaries and these groups differ in different countries, in differentiations. The races that we have in America are not the races that are found in Brazil, which are not the races that one might find in Cuba or in Mexico. This is the construction of a society.
As I said, racial categories differ with different societies and nations. Racial categories are fluid and situationally variable. I may identify myself as a person of one race in one context, but may identify myself as a person of another race in another context. When I am with my family, I may identify myself as a particular race; when I am out in the world presenting myself to the public, I may present myself as something differently.
I don't know whether you have all head this concept, but coming from New Orleans, this is a really familiar concept, the concept of passing. When you are assigned to be in one racial group, but because of your phenotypical characteristics, because of your appearance, you may successfully convince others that you are from a different group. This happens all the time.
There has been considerable research looking at multiracial adolescents to show that this is very context dependent and very situationally defined. It is contested by persons in the populations involved. Just because I say I belong to a certain racial categories does not mean everyone around me will agree to that. It is something that is negotiated, a reality that is negotiated.
However, that is not to say that there is no such thing as biological variation among humans. Of course there is. We look around and we see that there is differentiation. Probably a better way to explain this is what is called a cline. The American Anthropological Association has done quite a lot of work on this. They are actually having a national conference at the end of this month on race, ethnicity and emphasis upon clinal populations.
A cline is a single trait genetic gradient, not a group, but a gradient over space and different environments. So if you can see over here at site A, this is an example of skin color. The genetic material necessary for melanin production in site A is quite heavy over to the left. That population is adjacent to another population, and adjacent to another population, and on and on and on until you get to population N or site N.
What happens is that one group mixes genetic material with another group and with another group. It becomes more and more diluted over space and also with environmental variation, in this case, ultraviolet radiation. So where there is more ultraviolet radiation, there is a selective effect that is adaptive for more melanin production. So what you have here are not different categories of people; you have got a real gradient.
This is a cline map that has been put together by Nina Jablonsky and George Chaplin at the California Academy of Sciences. They have done quite a lot of research on distribution of various genetic traits across the world. The thing to look at is not the yellow bands versus the blue bands, but look vertically or north to south. You can see a gradation in colors. What you see is a gradation in skin color, gradient in skin color. You see clinal variation, and that is associated with the concentration of ultraviolet energy on the planet. However, it is not completely correlated with ultraviolet radiation, because there are other factors besides just that factor which come into play to bring about skin color. So ultraviolet radiation is obviously not the only variable there.
Anyway, so there is that map for skin color for a cline. You can take another map and map another trait, let's say waviness of hair, and superimpose it over there. The cline for wavy hair would not be the same as the clinal map for skin color. You could then superimpose another map for another trait, and on and on and on, until you would have tiny breakdowns across the globe of small populations that might share genetic material, but nothing like what we consider race in this country, or another country as Brazil.
So the point here is that there are no hard definitions, there are no clear boundaries. What we have are gradations of genetic material that vary somewhat with the environment.
Let's go to ethnicity. Ethnicity on the other hand, is the quality of being from an ethnic group. An ethnic group is a subcultural group of a multicultural society. You have to have a multicultural society, because if you only have one culture, how could you have ethnicity? You have to have two groups to have ethnicity.
Usually this is based on a common national or tribal heritage. I just want to make a little side track, and I know I am probably going on, so I am not going to say much about this. The point about culture is, despite what 19th century anthropologist colleagues and other people said, culture is not inherited, culture is learned. It is not genetic. It is passed on from generation to generation by teaching. Cultures shapes experience of the world around us, and it also changes over time; it varies. So that complicates things as well.
These are some of the things that comprise culture. We won't go into that.
The point with ethnicity is, again it is hard to pin down ethnicity. Is ethnicity based upon one's heritage, upon a shared experience, upon group membership? It is not necessarily geographically delineated. You can think about the Jewish Diaspora all over the world, so ethnic groups are not tied to a specific geographic location. Again, as race, ethnicity is situationally variable. On St. Patrick's Day, everybody becomes Irish. In different contexts, one emphasizes one ethnicity over another. Again, just as with race, this is contested. I might say I am Irish, but my family sure isn't going to buy it. I am French Creole.
Race versus ethnicity. I guess this is the comparison to be made, that race is a social category based upon a similar physical appearance, whereas ethnicity is a social category based upon either a shared culture or a shared cultural heritage, because you cannot assume that everybody who is in an ethnic group shares the same culture. Views upon this definition of ethnicity and how to draw the boundaries between categories differs with different researchers, different theorists.
Another point to be made is that race does not equal ethnicity. A racial group is not the same as an ethnic group. Any given racial group is not necessarily of the same ethnicity. Those people we call whites encompass quite a number of ethnicities. An ethnic group is not necessarily all the same race. Any given ethnic group is not all the same race. For example, Cubans.
I'm going to go on to this one. For example, for a little graphic emphasis, look at all the people that we would consider white, Georgian, as in the Republic of Georgia, not the state. I don't know whether the state of Georgia has its own ethnicity. Italians, French, Greeks, Iraqis, Swedish, Spanish, all these people look different, but we consider them white race. They are definitely different cultures
For Cubans, they include at least blacks, whites, Nestizos, and I know from my Cuban friends, do not attribute the wrong race to them. Do not say they are one thing when they consider themselves another thing. However, they will all, all agree to being Cuban. They will all present their ethnicity as Cuban. So it is really not the same. These two are different things.
Multiracial persons. In the Census 2000, there were over six million people who on the Census said they were two or more races. That is 2.4 percent of the population. But what is wrong with this figure? What is dodgy about this figure is, how do we measure race? We just considered all the problems of delineating race in everyday life. Do we consider primary race, one's primary identification to be what we will measure? Should it be how somebody else attributes race? Should it be the person who is doing the interview? Should it be somebody in the household, which is how the census does it. What about contextual changes? If you get a person in one setting, they may say that they are of one race on a survey, but if the survey is taken in a different setting, in a different context, you might get quite a different answer. So this complicates things.
Problems measuring both race and ethnicity is that both are subjective and both are contested. The question is whether self report is the way to go, whether ID by others is the way to go. It is context dependent, very situationally. Some studies have been done by Robert Hahn from CDC that shows that your race can change over the course of your life from your birth to your death records. He did an examination of birth and death records of infants who died soon after birth. So it is not a question of the wy things were measured over several decades, the same time period. Both records on one individual said one race, death certificates gave a different race for that same infant. Again, there are no clear boundaries.
Now, David Harris who is from the University of Michigan, he did a study with college students, showing them pictures and asking them to identify the person's race. What he found was that how one identifies another person's race depends upon the person's characteristics.
But what I want you to look at here in particular is, what race would any of these people be? Would you feel comfortable? So you can see the margin for error if somebody is in a clinical setting, putting down the person's race and medical record. You can see the problem with attribution of race by others. I don't think I need to say anything more about that.
Another problem is the question of lumping versus splitting. The OMB Directive 15 categories. There is the categorical use of non-Hispanic white, Hispanics, American Indians, Alaskan Natives, Asian Americans and so on. But look at the number of groups that are subsumed under those . I'm not suggesting that we measure for the Census the 600 plus groups of American Indians and Alaskan Natives or the 37 plus groups that might account for Asian Americans, the 20 plus that could be considered Pacific Islanders or over 100 languages. However, when you do the lumping, you lose some information, obviously.
Here is an example. This is from the 2000 census. Let's look at Asian Pacific Islander, the pink bar in the middle. According to this, Asian Pacific Islanders, less than ten percent are in poverty. However, if you break down Asian Pacific Islanders into their subgroups, you can see that quite a few of the subgroups are much more than ten percent in poverty. So you lose that kind of information by lumping.
I guess the question would be, or the guideline would be, what kind of data do you want to get and how are you going to use this data, as to whether you are going to lump or to split.
The questions of usage of race and ethnicity in research, can we use them as a proxy for other phenomena? They are used that way very often, but would it be better to just measure the phenomenon you are trying to get at, if possible? Can we really use race to research biogenetic phenomena? I think I have showed on what shaky ground we are on when we try to do that.
Is race the best variable for social, cultural and behavioral phenomena? I'm not sure about that. Certainly not the cultural, because we have shown that there are quite a number of different ethnic groups that make up a race. However, measuring race does say something about one's position in society and one's position in a power hierarchy.
Race can be used to study for policy purposes or studies of racial biases, which I think we will agree are very real. Ethnicity is probably more useful in understanding behavior, because it implies cultural influence. But why should we collect data in terms of race and ethnicity? It is associated with differential health risks and resources, both of these phenomena, to inform policy, to understand social and behavior phenomena related to health, and to track and understand bias and disparities relative to health.
This is going to be my last point, that it does make a difference in health. There is a study that probably many of you know, that is done by Kevin Schulman and a host of others, that shows the effect of race and sex on physicians' recommendations for cardiac catheterization.
What was done is, a sample of 720 physicians were shown these pictures and these pictures. They were shown videos of these people being interviewed. These people gave the same exact histories, complained of the same exact symptoms. Women were far less likely to be diagnosed with coronary artery disease, and women and blacks were far less likely, statistically significantly so, to be recommended for cardiac catheterization.
So to say that we don't need this information I think is really a mistake. A study like this shows how real the effect of race and ethnicity, however socially constructed they are, how real they are and what an effect they have in our society, and how important it is for us to collect this material.
Take-home message: Race and ethnicity are not the same thing. They are important for data collection, because race and ethnicity are significant in health and society, but there are measurement issues that must be resolved for adequate data collection and ultimately, if we are going to monitor and improve population health.
So there you have it. There is everything you need to know about race and ethnicity.
DR. MAYS: Thank you very much.
DR. LUMPKIN: Thank you. Personally, I'm a lumper.
DR. MAYS: On that note, I think we should turn to Dr. Cain.
Agenda Item: Race, Ethinicity and Socioeconomic Position - Virginia Cain
DR. CAIN: Thank you for having me here to talk a little bit about socioeconomic position and health today.
What I want to do is try and go over through a few examples the effect that socioeconomic position has, and where we are going with research in this area. From an historical perspective, I think the 1950s were a decade where there was a predominance of the biological and medical model. Increasingly, health has been viewed as a function of social, psychological and behavioral factors, and that comes together in socioeconomic position.
Socioeconomic position has been defined by Lynch and Kaplan as the social and economic factors that influence what position individuals and groups hold within the structure of society, i.e., what social economic indications of location in the social structure that may have influences on health. As I said, what I want to do is take you through some examples and talk about how socioeconomic position has been measured, and how that does actually affect the health of individuals.
The first slide I want to show you just touches on the race and ethnicity issue. And of course, after Suzanne's discussion, I'm afraid to use any of these categories that we all have our data collected in. But it is clear that there have been dramatic improvements in health, but not everybody has shared equally in those improvements, and you can see in this case that life expectancy for everyone has gone up among the black population; it has not reached the level of the white population. And in terms of socioeconomic position, race and ethnicity is one of the categories that can be used to define that position.
I wanted to talk a little bit about socioeconomic status. That is a term that has been traditionally used to capture some of the social and economic variables of interest. Socioeconomic position I believe is a little bit broader, in that it captures more than what we have traditionally used as socioeconomic status variables. I'll go through and explain a little bit more what I mean by that.
There has been a long tradition of socioeconomic status research, approximately 150 years of data that link SES to health. It has been linked in almost every health measure that you can imagine: All cause mortality, chronic and communicable diseases, injury and adverse effects, health related quality of life. It goes on and on -- low birth weight, infant mortality. So it has been a very powerful indicator of health outcomes.
How has it traditionally been defined? The most common measures that have been used are income, education an occupation, or some combination of those three. I'm going to take you through and show you some of the factors associated with each of these, to try and get across the importance this has on health outcomes and how strong these relationships are.
The first thing I want to talk about is education and mortality. It is often used because it is readily available. It is something that we can ask people and get pretty good answers. You don't see reverse causation so much in this. The education is established early in life. It is pretty stable beyond adulthood. As I said, its measurement is practical and convenient in many contexts, and it is a socioeconomic indicator that can capture a variety of aspects of lifestyle and behavior.
It clearly measures something important, in terms of health outcomes. This particular slide using NCHS data looks at the deaths per 100,000 population by educational attainment. It is clear, not so much a trend over time, but that the highest educated, 13 years plus, have the lowest death rates, and that has been consistent over that period, and I would say it is pretty consistent over the long term.
It also holds true for both men and women. You will see the male in the red on top; the relationship is more pronounced for men than for women, but clearly the higher levels of education have lower death rates, and that is pretty much a linear relationship.
It is also reflected in terms of the infants in the population. This is divided among racial and ethnic groups, starting with non-Hispanic whites, non-Hispanic blacks, Asian Pacific Islander, American Indian and Hispanic. If you look at these data, you can see that within every racial or ethnic group, clearly the highest levels of education have the lowest infant mortality rates. While there are striking differences across the different ethnicities, it is clear that in every single one of them, low education ends up leading to higher infant mortality rates, or is associated with higher infant mortality rates.
Education is an interesting variable, because the pathways and the mechanisms by which that is translated into health effects is not really very clear. That is one of the things that at NIH we have been trying to explore a little further to understand what it is about education that results in the dramatic health effects that you see. It doesn't seem to be that it is particularly education on specific health topics, but rather as you see here, it is simply years of education. So we are exploring how that translates into the health effects that you see.
I next wanted to move to occupation. Occupation has been used for a long time in looking at socioeconomic position or SES. Occupational prestige scores have been widely used. Clearly people with -- if you want to say more prestigious occupations tend to do better in terms of health outcomes.
I wanted to bring data from an interesting study that Michael Malmut has done in Great Britain, a very well-known study, the Whitehall study. What is particularly interesting about the Whitehall study is that it is a study of British civil servants. He measures a variety of variables among civil servants at all levels of occupation within the civil service.
It is interesting particularly because they are all employed, so employment isn't an issue. They all have access to the same health care programs, so that is controlled for. But even with that, there is a very clear relationship between health outcomes and occupation. In this case, the administrative or managerial positions would be considered the highest in the occupational ratings, followed by professional, clerical and others, which would include manual labor. It is very clear that the employment grade and coronary heart diseases deaths are related. The administrative positions are much less likely to have deaths than those in the clerical or other positions.
He took this a step further, to try and figure out what exactly is it about these kinds of positions that would lead to differences in the health outcomes that we are seeing, and explored some of the psychosocial factors that are associated with the different kinds of work conditions.
The best health outcomes or the lowest rates of heart disease were found among those who had fairly high job control. It increased among those who reported low control on their jobs, low control about how they spent their time, about their decisions that they made. The highest heart disease was found among those who reported that they had to engage in very high efforts, and there were very low rewards associated with the job. Clearly, that is a measure of stress of some kind that is associated with the occupation.
Income is a very important measure. It represents the extent to which people can control and have access to a variety of resources, life changes, in terms of, they are able to pay for additional medical care, pay for education, so it really is a very important variable when trying to assess socioeconomic position. And it is an important variable with regard to health.
Here are just a few examples I want to go over that show the relationship between income and health. Here we have several racial or ethnic categories, but within each of those categories the poor are strikingly more likely to report fair or poor health than those in higher or middle categories.
So among the -- just taking a look at the men, because the pattern is generally the same for both men and women, among the high income groups, less than five percent report fair or poor health, but when you get to the group which is characterized as poor, nearly 30 percent of the Hispanics, over 30 percent of the white and over 30 percent of the blacks report fair or poor health.
You can also see that in terms of reporting limitations of activity. There is a relationship here for both blacks and whites, with the non-poor being much less likely to report any kind of a limitation on activity, and in the case of the poor, almost 30 percent report some kind of limitation on activity.
As I mentioned, these are pretty traditional measures, and we have been using them for a long time, and they have provided some interesting data and have provided some suggestions for interventions and how we might improve health. But there has been in recent years additional conceptualization of what it means, what socioeconomic position means.
I'd like to talk a little bit about wealth, which is still an individual level variable, but it provides information about the extent to which individuals or families are able to have a buffer against the social or economic stresses that may occur, something like a period of unemployment, a divorce or even a positive case of a birth of a child still can be both an economic and a social stressor.
There has been some data in the past several years that has assessed the wealth of various racial and ethnic groups in the United States. Some of the data that have come out show that for every one dollar of wealth of a middle-aged white household and African-American household has 27 cents. So you might say, well, it has to do with the distribution of income, that African-Americans generally would be paid less, but taking this a step further, Jim Smith was looking at within various income quintiles, the distribution of the wealth. Even at the top quintile, African-American households have 56 percent less wealth and Hispanics have 67 percent less net worth than white households.
It is even greater in the lowest quintile, where African-American and Hispanic households have 85 and 63 percent less net worth. These income measures or these wealth measures include savings or whatever, but they also include real estate equity. so to the extent that people own homes or property, this is included in that as well.
When you take out the home equity, because that really is not very liquid, and if you suffer some kind of an income shock or some kind of a social stress, that is hard to get a handle on, you take that out, you see the dramatic differences in the median financial assets available to families of different ethnic groups. White households have over $17,000 worth of assets, African-American households have $400, and Hispanic households on average have $78. So essentially, there is no cushion for many households. There is -- should they suffer some kind of a social or economic shock, there is nothing to buffer them against financial or social -- changes in their financial or social position.
Over the past several years, people have been looking at communities and neighborhoods and the characteristics of those communities and neighborhoods, believing that they have some characteristics of their own that affect individuals above and beyond a sum of the characteristics of the individuals that occupy that neighborhood.
The kinds of things that people have been looking at are median or per capita income. So the analysis would be looking at whether the income of the neighborhood has an effect on the individual, the average income of the neighborhood has an effect on the individual over and above the effect of their own personal income or their own family income.
Other kinds of measures, percent in poverty, median level of education of the neighborhood or whatever the geographic unit, percent white collar occupations, unemployment rates, social cohesion, and the kinds of measures that -- units of analysis that these measures are taken on include census blocks and tracts, postal codes, metropolitan statistical areas, states, counties, and in some cases there are smaller units, geographic units, that are employed. I think part of the issue is a definitional problem in terms of what exactly is a neighborhood, what is a community, how do you draw those boundaries. That is under further investigation.
I wanted to just mention one example of a neighborhood effect. There are many, but this is a particular one, looking at poverty in Alameda County. It was associated with a 50 percent increased nine year death rate. That is controlling for person's own income, the whole variety of individual social factors that we have been talking about, and simply living in an area of high poverty resulted in a 50 percent increased death rate. Currently, there are studies looking at infant mortality, child development, life expectancy, a whole range of other health outcomes that are important and that could be affected by the neighborhood or the community that families live in.
How does socioeconomic position affect health? There are a variety of ways that it can affect health. It affects everything about how the choices a person has in terms of health care, education, access to information, nutrition,housing, where they live. We talked about some of the psychological pathways that result in stress in terms of occupation, self efficacy, the extent to which they can change their environments or change particular factors in their world, behavioral norms. If you look at what is normative in terms of some of the health behaviors that we know are very important in health outcomes, there are considerable differences by socioeconomic position. And it affects the environmental and occupational exposures.
It is a very complex situation, but what this does is, it provides a variety of places for opportunities for intervention. This is a multilevel model of intervention that Lynch and Kaplan have proposed, and I think down near the front of it, you see some of the traditional things where we have interpreted genetic factors, individual risk factors, pathophysiology pathways. In the health care community, those are pretty traditional areas for intervention. Clearly, people exist within a much broader social world, which can be hard to capture sometimes in terms of collecting data, but it provides opportunities for us to act and influence people.
What I thought would be perhaps a helpful thing to end with is to talk a little bit about where we need to go in the future with some of these data needs. I title this data needs, but in some ways it is data and research needs.
Clearly, longitudinal data. I don't know that we need to talk too much about that. But life course measures are very important. We capture things at one point in time, but it is frequently the accumulation of the effects that socioeconomic position has over the life course that ends up in the outcomes that we see.
I think we need additional, more sophisticated measures of socioeconomic position on an individual basis. I talked a little bit about expanding beyond income, education and occupation, but I think that there are other areas that we can move to and better describe, where a person sits within society.
I think that there needs to be more and better measures of community level variables. We have started approaching that. There are a number of different ways that community level variables can be collected. Some of the pre-existing data sets, some of the researchers interview community members about their communities, some approaches have been to go into communities and count various factors like broken windows or other physical characteristics of the community. I think that there is room for a great deal of work in developing those measures better.
I think that we need to come up with -- this is maybe a conceptual issue, but to come up with additional ways of integrating the variety of measures that I have talked about and that we may develop to determine socioeconomic position.
So I'll end there. Thank you.
DR. MAYS: Thank you very much. Let's take some time for questions, comments from people.
DR. LUMPKIN: Jeff.
MR. BLAIR: I guess I have got so many questions, that I will -- there is a hard copy of the report that I will be able to take home and two weekends from now I will be able to read it over the weekend. Your phone number is there where I can call you back. Thank you, though.
DR. LUMPKIN: Mark.
DR. ROTHSTEIN: I just want to follow up on that. Are we going to be able to get hard copies?
DR. MAYS: Everybody is asking for them.
MR. BLAIR: The essence of the questions that I would have is, this is wonderful background information, and it is interesting. I guess I am groping at the end of the presentation to see how it could relate to the work we are doing on NCVHS.
I am for the most part on Standards and Security and NHII, so some of the questions -- and you probably won't be able to answer these, but are the kind of questions that start to occur to me is, if we try to hit low-hanging fruit and try to solve the disparities most, if we had universal access to health care, how much of the problem does that solve. Then if there is a study that says if we did have that, then what are the remaining disparities, and how does that deal with an information infrastructure.
So those are the host of questions that come to my mind. So it may be a next level.
DR. LUMPKIN: Mark.
DR. ROTHSTEIN: I have another question. I thank you both for your presentations. This question goes to Dr. Heurtin-Roberts. It deals with pharmacogenomics. I saw a TV commercial within the last week for some pharmaceutical product, and I don't remember which one. I think it may be an anti-hypertensive, but it really doesn't matter. After talking about how great a product it was, they discussed the side effects, and at the very end, they said side effects are more pronounced in African-Americans.
Now, this of course is a crude and inaccurate statement. To be accurate, the statement would have had to say, side effects are associated with a particular allele or group of alleles that due to ancestral patterns of migration, founder effect, genetic drift and other principles of population genetics, have become associated with what we socially construct and individually self identified as African-American populations.
So nobody would understand that, of course.
DR. LUMPKIN: Except other lawyers.
DR. ROTHSTEIN: So the question is, we have got two equally unsatisfactory choices. One, we can use the technical descriptions which very few people understand, or we can use these very crude lay descriptions that not only misrepresent the science, but reify the misconception that there is a biological basis for what we call quote race. I am wondering whether in your research you have come up with any insights as to how to get over that problem.
DR. HEURTIN-ROBERTS: No, but I have some thoughts about that. One of the things that needs to happen is, there needs to be much greater education of the general public about what race means, about the meaning of race, and what an imprecise and not biologically driven categorization scheme it is. That is going to take time. That implies social change.
I really don't have problems with people saying that this group tends to, or that there may be. What I am concerned about in pharmacogenetics is the ecological fallacy, wherein because you are faced with one person -- because you know about a larger population, what you may know about a larger population, and you are faced with an individual in a clinical setting, that you may not give that individual the individual regard they deserve. It is much easier to say you seem to be part of this group, so this is how we will treat you. That is the sort of thing that I am worried about.
I don't have problems with that sort of thing, but I do wish that there was greater awareness on the part of the public.
DR. ROTHSTEIN: You can see where it tends to --
DR. HEURTIN-ROBERTS: Yes, sure, it does reify things, yes. I do think that one of the things that needs to happen is, we need to have these conversations much more, not only in this group, but the pharmaceutical industry as well, with other researchers.
DR. LUMPKIN: But it is one thing to expect the public to understand that, when the public doesn't even understand what genetics are.
DR. HEURTIN-ROBERTS: Right.
DR. LUMPKIN: But it is amongst the scientific community and others who have not yet fully understood that race is not a genetic concept, where it is very important for these issues to be discussed, because there are a lot of assumptions that are made, and people who don't have a scientific basis in the field, that they are making assumptions about.
DR. HEURTIN-ROBERTS: The other thing that disturbs me is the marketing towards a particular group that the industry might engage in. That is based on incomplete and inadequate information, yet because of a perceived market, the industry may take a certain marketing perspective.
DR. LUMPKIN: Mike.
DR. FITZMAURICE: I was really impressed with the framework both of you presented. It is clear that there is not enough data or not enough access to data to let us know more about these characteristics, about how the population characteristics are measured and their effect on health status, among other quality of life measures.
As we move to what social policies should be developed, given what meager information that we have, I am wondering, is it wise social policy to try to change the characteristics so that by changing the characteristics you lead to better health status -- I suspect that perhaps not -- or is it better to compensate those in lower categories by taking from those in higher categories, is it better to try to change the structure of society with education, move some of these people from lower job classifications to the professional and administrative jobs, better to try to reduce discrimination, that deprives people of what they might otherwise have based upon the merits of what they can produce? Is it better to try to change the physical location from one environment to another? Is it better to effect a cultural change?
As I look through the presentation, I am trying to look for the social action levers and triggers, and to see what additional information you have given us that we might advocate that we need more information in these particular areas, because they can effect a change in health status, to use an example. But it is also income status, it is quality of life, and it is being happy with who you are and what you are.
DR. HEURTIN-ROBERTS: I would say that there are no easy answers. There is no one perspective, there is no silver bullet. It is a complex problem and it is going to take complex approaches, I believe.
But I do think that the work we are doing in the Subcommittee on Populations would do -- we are discussing a lot of these issues, in trying to prioritize some strategies to recommend to the larger committee and to the Department. But again, to say one is better than the other, more research needs to be done, and no single strategy.
DR. CAIN: I would say both, obviously, but in some ways that is an easy answer. But I think both needs to be done. Even if we were to all agree that societal change were a good thing in many of these cases, that is not going to happen overnight by any means.
One of the things that we try to do is to try and understand what it is about some of these factors that are causing the differences that may lead to an immediate intervention. For example, in education and health, where there is the strong relationship, we really don't know what it is. Just having an extra year of education, you wouldn't think would have -- what is it that causes the differences in health outcomes? Is it how people interact with the medicare care system? Is it how they understand instructions that they have been given. What specifically is it about that?
In some of the grant programs we have, we are trying to look at what those mechanisms are. So if it is that they need training or the health care providers need training on how to interact with different people, then that is an intervention that we can start to work towards without raising everybody's education level.
So it is really trying to tease apart what are some of the factors that cause these relationships, the pathways and mechanisms, and see if we can intervene there.
DR. FITZMAURICE: Are there particular data sources that you would recommend to us that we go after to learn more? For example, we need more access to IRS income data, we need more access to health data? We would have to consider the privacy aspects of that, of course, but --
DR. CAIN: I think that is one of the issues, is linking data sets, that can be very important to try and -- because one of the things that happens frequently are that the good health data sets don't have sufficient social and economic variables on it to do some of the kinds of analysis that we would like to do, and the good social science data sets don't have good health information on. So what people often try to do is link up various data sets. As you have pointed out, that can lead to privacy issues.
DR. LUMPKIN: Charles.
PARTICIPANT: Since much of the data come from my area, I would like to comment on some of the problems that we have, and commend both speakers for what they have done.
Let me give you some hard examples. We had an uptick in infant mortality for the first time in some 40 years. Now that we are really looking at what the causes of that by disparities, we are looking not at the infant mortality record, but the linked infant birth and death record, because the information on race on the birth certificate is superior to that of the death, because it comes from the mom on the birth certificate. It is more than is reported either by the physician, the hospital staff or the funeral director on the infant death. So that gives you a better look.
But then, is the birth certificate all that good? The birth certificate is a legal issue, so what is that mom thinking or whoever that is who is filling out that information at the time? For example, in the late 1970s, there was a terrific -- a fairly significant influx of middle-aged American Indians in North Carolina wanting to change their race to American Indian from what it was before. Why was that? It was because there was a new program in place that might help them out. So the legal implications had something to do with their reporting of who they were at that time.
We are going to be coming out with a report -- not we, excuse me, one of our analysts in Demography next month, which is going to look at Hispanic differentials, the Hispanic protection in mortality. It is going to basically say that that is flawed. The reason why it is is because we have linked up to other surveys which in fact allow us to look at the true race, we believe, and see that protectiveness that we thought just through looking at mortality data is not in fact there.
So why am I saying this? I do think that statisticians have very much of a responsibility to say the strengths of our data and the weaknesses of our information as we report it. I think we also have to improve our research on what we should be collecting, relative to socioeconomic variables, to tease these problems out. Sometimes I think we make the situation worse in what we report.
DR. LUMPKIN: At this point, we are actually 15 minutes over our allotted time. This is obviously an issue which is extremely complex. Let me just throw one variable in there to demonstrate that.
When you deal with health disparities, it probably doesn't matter what the true race is; it is what the health care provider believes that race to be, because that influences their decision making. So you have the complexities of even whether or not you use self reporting, which is considered to be the gold yardstick.
So we have a lot more to discuss. This really is an opening salvo of our discussion. I think the message from this discussion is that we need to discuss this more and have additional sessions at future meetings. I'd like to thank both of you for coming. Thank you.
Let's take a ten-minute break, and we will get back at 25 after.
(Brief recess.)
Agenda Item: Reports from Subcommittees and Work Groups
DR. LUMPKIN: We have our DFO, designated federal officer, official. We have a quorum, so let's reconvene. We have a set of reports from subcommittees and work groups. We are probably going to hit the letter on e-prescribing a little bit later.
We have had a report from the executive subcommittee. The NHII work group report. We met yesterday. We are fleshing up our hearing on November 12, which we believe will not be the only hearings that we hold on the personal health record, so more to come on that. We have a conference call coming up on September 20, and we will continue to plot out our agenda. As Simon pointed out, the agenda is getting full enough that we probably won't get away with one-day hearings anymore in the near future.
Any other questions or additional items on the NHII?
DR. GREENBERG: I think we can expect Dr. Braylor or someone from his staff probably to join us at the November meeting, because they weren't able to this time.
DR. LUMPKIN: Good. We are going to skip over Standards and Security to go to Privacy and Confidentiality.
DR. ROTHSTEIN: Thank you. Just some announcements about our plans before we get to the revised letters. We have hearings scheduled November 18 and 19, one half day on security and privacy issues, and a full day on the issue of e-prescribing and privacy. We are in the process of planning a mid-January meeting sometime next week. Staff will call the subcommittee about available dates.
Let me review yesterday what we did. We approved the planning letter and the marketing letter, and the subcommittee was directed to make some revisions to the fundraising letter, which we did, and which we have distributed to you.
John, I don't need to read that, correct?
DR. LUMPKIN: Jeff, have you read that?
MR. BLAIR: I haven't read this last version, but I was there when they were discussing it, and I feel satisfied you don't need to read it for me.
DR. LUMPKIN: Okay.
DR. ROTHSTEIN: So let me see if I can call attention to some specific parts that are of particular interest to the members of the committee.
If you will take a look at page two, there are three things I want to highlight for you. The third paragraph is a new paragraph to address the concern that there was no mention made of the consumer viewpoints in this. That summarizes the written testimony from the Georgetown University Health Privacy Institute.
We have broken down the recommendations into three bullets. I want to highlight two things related to those. In the second bullet we put in parentheticals after the description of broad designations, for example, surgery, oncology, but not narrower designations, to give a greater indication of what we have in mind by department of service Information, and we added a third bullet which explains our theory of how this fits in with the overall scheme of things.
Jeff, it says, the covered entities' notice of privacy practices should inform patients that their department of service information may be used in fundraising, and patients should be afforded the opportunity to decline to permit their department of service information to be disclosed for fundraising or to opt out of all fundraising contacts.
I thought for the benefit of the members of the full committee, I would go over what I see are the practical implications of this, what this does to the current state of the privacy rule and how it relates to other things we are doing.
First, what this recommendation would do is to go from an opt in to an opt out system. Currently you have to opt in to disclosure of this kind of information. Now we would make it an opt out.
Second, patents who don't opt out -- sorry? It is currently an opt-in system.
DR. HOUSTON: Oh, for fundraising for additional information, not generalized fundraising.
DR. ROTHSTEIN: Correct.
DR. HOUSTON: Just to make sure that is clear.
DR. ROTHSTEIN: So in terms of department of service information, it is currently opt-in, and we would go to an opt-out system.
The second point is that patients who don't opt out will have only broad designations disclosed, such as surgery patient or oncology patient, but not anything that would reveal a diagnosis or a particular provider.
Third, patients will be able to opt out of any contact as well as department of service information, an option they currently do not have.
Fourth, patients will have a second opportunity to opt out after the first contact. Currently, the rule provides that only after the first contact can you opt out of further contacts.
What are the relations between these recommendations more broadly? I think that the intent of this -- and will succeed in equalizing the disclosures for fundraising between specialty hospitals and comprehensive medical centers. That is one of the specific intents. In addition, in response to a question that was raised yesterday, I believe it harmonizes our position expressed in the marketing letter with our fundraising position. In other words, there is in the marketing letter a recommendation that the Secretary explore opt-outs or additional contacts for treatment information. Here, we are also recommending an opt-out procedure for both broad designations as well as all contacts for fundraising. So we are recommending opt-outs in both situations.
So with that, I think I would like to entertain comments.
DR. HOUSTON: There is one other point that I missed when you said it. There is going to be an explicit recommendation that the notice of privacy practices, that don't need to necessarily have a notice -- within the notice of fundraising opt-out, the recommendation would be that it also be a requirement of the notice of privacy practices for experiments that decide to do fundraising.
DR. ROTHSTEIN: Yes, that is certainly correct.
DR. HOUSTON: I don't think it was clear that the status quo today is that it doesn't have to be part of the notice of privacy practice. So this would be an additional protection. I missed it when you were talking about it.
DR. ROTHSTEIN: Yes. In the third bullet, to repeat, we now add, the covered entities' notice of privacy practice should inform patients that their department of service information may be used in fundraising, et cetera.
DR. LUMPKIN: Any questions? Jeff.
MR. BLAIR: I move that the letter be approved.
DR. LUMPKIN: Moved by Jeff, seconded by Mark.
DR. ROTHSTEIN: Seconded, if permitted by the rules.
DR. LUMPKIN: Sure. Actually you could have moved it, but Jeff was quicker. All those in favor, signify by saying aye.
(Chorus of Ayes.)
DR. LUMPKIN: Opposed, any? Great . Anything else from Privacy?
DR. ROTHSTEIN: No. Well, I have one thing that comes under future agendas, but as long as I have the mike now, at the November meeting of the full committee, we would like to request staff not to put the subcommittee meeting for privacy and confidentiality up against the populations subcommittee, because we are planning to do something jointly with them.
DR. GREENBERG: Should we plan a joint meeting?
DR. ROTHSTEIN: We want separate meetings, but there is gong to be a presentation by Vicki and her subcommittee at our subcommittee, in terms of doing some joint things. So if we have separate meetings, I think that would be most --
DR. GREENBERG: The reason I asked, the problem that we are up against is that we also can't put privacy and standards at the same time, because there is so much overlap in membership, not to mention the NHII.
MR. BLAIR: Not try to solve it at this time, just as an issue that needs to be worked out.
DR. GREENBERG: Yes. I just wanted to mention that, so that you don't say we will have to discuss it. It is easier said than done.
DR. LUMPKIN: Are both the committees going to meet for the same amount of time? Subcommittees?
DR. GREENBERG: We can.
DR. LUMPKIN: So they will be separate but equal times.
DR. ROTHSTEIN: I suppose we could meet at the same time, and then just have an overlapping session.
DR. GREENBERG: That is what i would suggest.
DR. LUMPKIN: Okay. Subcommittee on Populations.
DR. MAYS: In the agenda book should be the letter that was completed, the commentaries. We wanted to thank you for your comments on that.
We do want to ask for one bit of help, which you can e-mail. When we talked about these issues before, there were a couple of people -- I was particularly hoping that Steve was here -- that said these are the kinds of issues which should be distributed more broadly. So now that you have the commentary, if there is any particular group or organization or somewhere that you think we should share the commentary that went to the Secretary, could you please e-mail me or one of the staff, Audrey or Dale, so that we can work on that? We are going to prepare a cover letter, and we are going to send this commentary out to what has been the typical individuals who have come to our hearings. But you all had some other suggestions that you were thinking about, so I really would like to be responsive to that, because I think that you had great ideas when you talked about it. We just needed specifics. So I would appreciate it if I can get you to follow up on that.
Work on our plate right now. We aren't announcing right at this moment all of our dates, but we are planning to have a meeting, because we have several things that we need to talk about. One of the things that we need to meet and talk about was -- the presentation that we had today was quite apropos of the issues that we are struggling with, with our colleagues in Quality. That is, there are a set of candidate recommendations in Quality that are in the auspices of population, in the sense of the collection of data on race and ethnicity, where that should be done, what is feasible to do. So we need to talk about what needs to be collected. Just saying to somebody, collect race and ethnicity is not as simple as it looks. Also, the whole notion of how do we get this into some of the mechanisms that are used, where do we do this. So we are going to have a two-day meeting, where we are going to be talking about that as a topic, so that before we start hearings, which we will do jointly with Quality, we know clearly what our question is and what goal we want to try and achieve.
We also should be coming back with our population report in November, so we back dated it. Since it is a report and it is fairly long, we want to try and get that out to you sometime in the end of November, if I remember. I may be doing these dates wrong, but the staff will help me.
DR. GREENBERG: End of September. No, excuse me.
DR. MAYS: Don't they get it the end of October?
DR. GREENBERG: They will get it at least a week before the full committee meeting. But it will go to the executive subcommittee prior to that.
DR. MAYS: So we are working on that. We are expecting a response -- and I am glad that you brought up the national child study -- we are expecting a response. We had them come in and do a presentation to the full committee, and we sent them a list of questions. So we are expecting a response back. After we get the response, there may be either some issues that we will discuss, or it may be something that we can bring back to the full committee, once we see what the response is. So that is on our agenda.
The other thing that is on our agenda is collection of medical history statistics. Many of you need to remember that Population doesn't just do race and ethnicity, we have a pretty full agenda in terms of being the Subcommittee on Populations.
We have put together a small task group that is trying to gather some information on what are the statistics that will be helpful for the federal government to collect on an ongoing basis on medical history statistics.
I think a big issue will be how we define medical history, because it can go all the way from just a focus on disorders to, how happy are you. We want to try and come up with what I think would be a good sense of what we think is necessary coverage of medical history statistics, as well as trying to probably focus on there being a relationship between data collected on physical health and data collected on medical history, so that it is a closer mind-body relationship. Usually the books that are put out are all physical health or all medical history, so I think we will be looking at that issue.
I don't think there is anything else. Dale, Audrey, am I missing anything? So that's it. We will have a pretty full schedule a little bit later after we have some planning time together.
DR. LUMPKIN: Thank you. Work Group on Quality.
MR. HUNT: The work group conducted two days of hearings back in June, on the 16th and 17th, to solicit comment on the eight candidate recommendations that were primarily on the Quality work group, along with the Standards and Securities work group, or subcommittee, excuse me from my linguistic failings, to see where we were.
At those two hearings, both Harry and Simon joined us, which was very helpful in illuminating the content. There is a further hearing scheduled on the 14th of September, where we will solicit inputs from the NUBC, NUCC, NC X-12, HL-7, about the mechanics related to the possible recommendations. So that will further the work group's understanding of what is going on in what we are talking about.
We had a lot of further discussion this morning around building our own matrix of understanding of what is it we are trying to do, what are the steps that we are talking about, and what is their difficulty. Part of what I think we are doing is rebuilding an institutional memory within the work group that we didn't inherit, except for John and Marjorie, and we don't always have the benefit of both of them. Usually we have one or the other, not always both.
The way this will play out, I believe, is that at some time in the future, not yet scheduled, we will come back to the full committee saying, these candidate recommendations now appear to be ready for full action and forwarding. That will also begin to talk about what do we do next as far as the Quality work group.
We had a brief discussion this morning, Vicki joined us for awhile, to talk about our overlap there. The discussion centered on that the past practice of the Quality work group doing a major part of its work with the full committee still seems appropriate. So there is no intention for us to go off and make big decisions, and come back with a summary. I think we will try to keep that decision pretty open and process it before the committee. So that basically means that we don't take six-year mouthfuls; we are going to have to shorten down to one-year mouthfuls that we can work on and make a manageable contribution.
Any work group members wish to add to that? Marjorie reminds me of my favorite issue, risk adjustment, which I have talked about, but we are not ready to say what that should be as far as the full committee. I think it will be an agenda on the full committee discussion, but there is some more work to be done within the work group when we are ready to propose that.
Agenda Item: Future Agendas for NCVHS Meetings
DR. LUMPKIN: Thank you. Let's move on to future meetings. Simon is on a conference call, so I am trying to give him time to return before we move on to the letter.
Our next meeting is 4th and 5th of November. Any action items that we foresee coming forward?
DR. GREENBERG: We have the Populations report.
DR. LUMPKIN: We are going to have a discussion on research.
DR. GREENBERG: Regarding action items, this report from the Populations subcommittee -- unlikely we will have anything from Privacy, because your meeting will have been just a few weeks before. Quality we will discuss after September 14, whether there is anything they want to take forward, so there is a possibility of a letter on one or more of the candidate recommendations.
DR. MAYS: Bob has suggested the American Community Survey.
DR. GREENBERG: Right now I am talking about action items. Are there any other action items that people are aware of?
MR. BLAIR: Is Maria or Karen with us? No? I think it would really be helpful to get some feedback from CMS in terms of how useful and appropriate the regulations are for e-prescribing, if it is possible for them to give us feedback. In a sense, it is a critique.
I know it typically goes in the direction that, we make recommendations, and we hope we are right, and we hope they are accepted. But in this case, there is such a collaborative involvement here, in terms of making recommendations that are actionable, timely, appropriate, meaningful, because we spent too much time on this. If we did not make useful, meaningful, actionable recommendations, I think it would be very helpful for us to get feedback on that.
DR. SCANLON: Yes, and I think on the assumption that the full committee will approve the recommendations in the next hour, we scheduled Simon to present the findings to the Data Council next week, so you will get some immediate feedback then. Then we will certainly see what we can do, Judy.
The thing is, I think we will be looking at these in the context of a proposed regulation down the road, so we will just have to see what the nature of the communication would be publicly.
DR. GREENBERG: More information items that we had planned for November, we will have to see if we can fit them all in, is the NIH road map, which was deferred from this meeting, so that had already been given a high priority by you, and hopefully the presenters will be available in November.
We also had a request for a presentation on the American Community Survey. It will be appropriate, since the BSC, the Board of Scientific Counselors, will have met to have a briefing from Dr. Robinson and Mays on the BSC meeting.
Another thing that had been proposed, we may have to defer this to a later meeting, but the clinical trial research agenda or infrastructure and CABG, not like in borscht. But there may be some opportunity to ask some questions about that at the NIH road map. Maybe we will have that presentation at a future -- I don't know.
Is there anything else? We had a bit of a discussion this morning about whether we wanted to have some discussion in the November meeting about risk adjustment. It may be premature.
MR. HUNT: It may not be ready.
MS. CARR: Were we going to come forward with the research topics from the subcommittee?
DR. GREENBERG: Definitely, each of the subcommittees or work groups has been asked to be prepared to discuss research topics. Probably what we will do is have that on the second day, so that there can be some discussion in the breakout sessions as well. So everyone should put that on their agenda.
MS. CARR: And is there any sort of template that folks should follow? To John's comments earlier, we want to pick something that is not being done already, but has significance, and in terms of outlining what we are going to do, incorporating that, that this is something that is not being covered, and here is what the benefit is.
DR. GREENBERG: I think there are two categories. One is, as I said before the break, I think it would be good if each subcommittee or work group could pull out their recommendations from the last -- even like two years, certainly because that is what this report is going to be, the 2003-2004 report, so that would be most appropriate. I know that there are some research recommendations in quite a few of those. They might be in all of them. So I think it is important to pull those out and talk about whether they need to be fleshed out more, whether we wanted to prioritize them or what have you.
Then areas that have not yet been identified, it may mean that some of them, you are prepared to say right now you would like to put forward as being recommended in the research agenda, but some of them may require -- although the National Committee doesn't do the research, it does often have hearings and exploration to determine where more research is needed.
So some of these areas may not be to go into the 2003-2004 report, but to do further exploration as to the type of research that is needed, maybe a followup on the type of discussions we heard this morning.
I think as John mentioned, it is not just the vision that is influential, but the more specific -- without micro managing, but the more specific the recommendations can be that are from the Department, the more people within the Department have to chew on and to say, yes, that is something we could do, or we are already doing it, or we need to do something different, but something to engage. If you make recommendations that more research on some very broad area, it is kind of like motherhood and apple pie.
So I think completely new areas that haven't been discussed might be fodder for hearings or for future deliberation. So I guess maybe we can send out a little bit more guidance this. But that is my off the cuff thinking.
DR. MAYS: Would it be useful at all to say which agency we think this is directed toward? Maybe CDC or NCHS or NIH? This is where I think we also get into pulling in these liaisons a little bit more. Like in POP, there are things that are census related, or there may be things that are Department of Education. At least if we are saying that, I don't know how it works for the report, but just in terms of for us, knowing that these are things that we are dialoguing with very specific groups on, would that be more helpful?
DR. LUMPKIN: I think we need to be a little bit careful. We also have to be concerned, as Phil Lee -- for those of you who don't know Phil Lee, the Assistant Secretary for Health, in one meeting he said, I always knew that data was a four-letter word, but I never knew it was spelled t-u-r-f. We have to be careful; there are different agencies that may consider that they have a role in that research, and we don't want to narrow our recommendations in such a way that we would particularly upset one agency over another.
DR. GREENBERG: What I'd rather see happen is to try to identify high priority research areas, and then engage with the Data Council members and others, and the Data Council can invite people from Census or others to come to those discussions, as to where are the possibilities for doing this type of research, maybe even inspire some cross-cutting approaches, rather than aiming -- if you have been working with agencies on particular areas, that can be mentioned, but I agree with John.
DR. STEUERLE: The topic of research is such a broad area. I can think of areas of research we haven't even mentioned here. I'm just wondering if there is not some more systematic way, if we are really thinking about what research agenda we want for the future, there is probably some more systematic way to ask ourselves questions about how to go through and filter.
It sounds right now like a process that is pretty random. We are all going to say something that is of interest to us, but there is probably some way to facilitate the discussion.
DR. LUMPKIN: Yes. What I am suggesting is that it be more focused on what the activities are of the subcommittees and the work groups, so that narrows it down, rather than trying to come up with a fairly visionary document on what kind of research related to health information policy that HHS should do.
So this research agenda should really be based in our current work agenda.
DR. GREENBERG: I think you start with the research recommendations that have already been made in these reports, and then if there are other areas where you are currently doing in your work plan, where it is obvious that there needs to be more research.
So I don't think it was the intention to be completely open ended, but it could be for starting to think about the future, areas that maybe need to be explored for developing a research agenda.
I think we can have more discussion about that in November, but I would think between the 2003-2004 report, it would be for areas that have already been identified or that you would like to identify from work you are currently doing.
Agenda Item: Letter to the Secretary
DR. LUMPKIN: Anything else on the upcoming meetings? The final item we have on our agenda is, surprise, surprise, a letter to the Secretary.
DR. COHN: That's right, a big letter to the Secretary. John, thank you for holding them off while I did my short lunch break conference call that turned out to not be so short.
I want to thank the subcommittee for their work starting at 8 o'clock this morning, actually 5 o'clock California time for the record, working on this revised document. I also want to thank Margaret for her work on giving us a relatively clean but subtly red-lined document.
The purpose of this conversation is not to go through every single word here, but to look at the changes that many of you suggested we make, that you think we have referenced relatively successfully. Some of your comments are in the body of the text, some of them turn out to be footnotes, so we will walk through and look, and if they meet your needs, we will move it forward for adoption. If you have further suggestions for improvement, we will talk about them as well. Okay, John?
DR. LUMPKIN: Great.
DR. COHN: So page one is unchanged.
DR. LUMPKIN: No, that is not true.
DR. COHN: I'm sorry, you're right. In the first sentence, it reads now, the NCVHS has been called upon by MMA to develop recommendations for uniform standards to enable electronic prescribing in ambulatory care. So in ambulatory care is the change. Thank you, John, I'm glad you have yellow-lined your document.
Page two is unchanged, as is page three. Page four, there is a slight change here that didn't quite make it, isn't well represented here. There is something that was supposed to be a bullet that is still in the text, if we can look down at guiding principles.
DR. LUMPKIN: No, it is a footnote.
DR. COHN: I may be looking at one of my 12 versions.
DR. LUMPKIN: Do you want my highlighted version?
DR. COHN: You need your highlighted version. We will make it through this one way or another. I'm glad this came before we made it through 15 pages.
It is currently down in the highlighted -- it now reads, These elements also align with the dimensions of care identified by the Institute of Medicine, Crossing the Quality Chasm, that care be safe, efficient, effective, timely, patient centered and equitable. So that is the addition based on one comment made during our session yesterday.
DR. LUMPKIN: We are going to walk through, so speak now, because once we move off of that point, then we are going to move to adopt the full document. So if you have any concerns, now is the time. Go ahead.
DR. COHN: On page five, we have added a sentence, and I will read the sentence. It is one of the transition paragraphs between types of standards and work that we did before observations and recommended actions. It says, The NCVHS notes there are privacy considerations uniquely related to e-prescribing, which will be investigated in subsequent hearings and reflected in March 2005 recommendations.
The good news is that when we came up with this, we had a quorum on the Privacy and Confidentiality Subcommittee with us, so they agreed that this was doable. The chairman unfortunately was not there, but he subsequently looked at it and nodded his head and agreed. We felt that this reflected a lot of the discussion yesterday related to, that there were potentially lots of privacy considerations that we needed to look at. So hopefully this adequately reflects the view.
DR. STEINDEL: I was just commenting to Margaret, the style we have been using, instead of the NCVHS, is NCVHS.
DR. COHN: Okay. Our next change is observation there, prescription messages. Remember, this was the beginning of the conversation on privacy that we had yesterday. What we have done is added a sentence and modified the last sentence. Let me just read it. This all relates to fill status notification.
The fill status notification function is not used today. Testimony revealed that there were questions about the business value and clinical utility of the fill status notification function, as well as possible privacy issues.
DR. VIGILANTE: Was it indeed never ever used? There is no single substantiation that it has ever been used? Or is it not widely used?
DR. COHN: You hate to say never as well as always.
DR. VIGILANTE: It is not widely used.
DR. COHN: We used currently under utilized for the cancellation and change functions. We could say little used today.
DR. VIGILANTE: Nobody said they used it.
DR. GREENBERG: Apparently it is not used today.
DR. COHN: We can look at what Margaret did and see if this needs -- I hate to say apparently.
DR. VIGILANTE: That is a bad word. I retract that.
DR. COHN: How about, is generally not used today.
DR. VIGILANTE: Reportedly not used.
DR. STEINDEL: Can we say something like, as far as the NCVHS has determined the fill status notification is not used today, because that is true. There was no testimony indicating use. I would vote for that. That is the factual statement.
DR. LUMPKIN: There is no testimony that indicated that the fill status notification is used today.
DR. VIGILANTE: I would say, to suggest that.
DR. ROTHSTEIN: But was there testimony that it is not used?
DR. LUMPKIN: Yes.
DR. ROTHSTEIN: Well, that is not what it says.
DR. COHN: Testimony indicated that the fill status notification is not used today.
DR. LUMPKIN: Testimony indicated that the fill status notification is not used today.
DR. COHN: Are we comfortable with that? As well as possible privacy issues. Are we okay? Our recommended action 3.2 has now been modified. The first sentence remains the same, where it says HHS should include the fill status notification function of the NCPDP Script standard in the 2006 pilot test. The new sentences now are, These pilot tests should assess the business value and clinical utility of the fill status function, as well as evaluate privacy issues and possible mitigation strategies.
DR. LUMPKIN: Very nice.
DR. COHN: I don't see any changes on page seven. Page eight, we have additional wordsmithing on observation seven, which has to do with prior authorization messages. I'm not going to read the entire paragraph, but it is the last two sentences, where it says it is estimated -- and this is just further explanation of the amount of prior authorization going on. It is estimated that two percent of prescriptions now require prior authorization, and that there is a disproportionately higher rate for Medicaid payors. This provides a natural body of experience from which to draw data for study of economic and quality of care impacts.
DR. LUMPKIN: This section was prior authorized by CMS?
DR. COHN: I think I will just move on to the recommended action here.
DR. STEINDEL: John, yes, it was, but it was using the incorrect 278 version.
DR. COHN: So we didn't get a response. Recommended action 7.2, HHS should support standards, development organizations and other industry participants in developing prior authorization work flow scenarios to contribute to the design of the 2006 pilot test. I think that was the only change in recommendations there.
DR. ROTHSTEIN: Simon, can you go back to observation seven?
DR. COHN: Sure.
DR. ROTHSTEIN: Do we need the word disproportionately?
DR. STEINDEL: That was the testimony.
DR. ROTHSTEIN: I don't know what that means. It is two percent across the board and presumably higher for Medicaid. Can't we just say that there is a higher rate?
DR. GREENBERG: Is it a considerably higher rate?
DR. STEINDEL: I think that's fine.
DR. COHN: I just want to make sure, the reason that we put that in was based on our conversation. It also reflects on action 7.3.
DR. LUMPKIN: HHS should evaluate the economic and quality of care impacts of automating prior authorization between dispensers and prescribers; 7.3.
DR. COHN: I just wanted to make sure this all hung together. This was additional commentary.
The next changes are in observation nine, clinical drug terminology. We changed a whole paragraph here, where there has been some significant changes. I will read the paragraph as it is being proposed.
The national drug code in NDC is used by dispensers to identify package drugs. However, NDC is not appropriate for use by prescribers in describing the clinical drug. To address this need, NLM has produced a clinical drug terminology, Rx Norm. Rx Norm provides links from clinical drugs to their active ingredients, drug components and most related brand names.
Preliminary analysis of the prescriptions written by the Department of Defense found that within the first thousand most common prescriptions written, the NLM matched 974 to Rx Norm codes.
Then we have, while comprehensive Rx Norm terminology for all marketed drug products, including generics, repackaged products and over the counter medications will not be available until the structured product label, SPL, for all those items that become available to the NLM from the FDA, the Rx Norm contains sufficient codes to be included in the 2006 pilot test.
DR. LUMPKIN: Eighteen is new, that footnote, by the way.
DR. COHN: Yes, 18 is new. I'm just trying to make sure the sentence reads --
MS. CARR: Simon, just as a point of information, Stuart added that phrase for some reason. Maybe Vivian would know.
DR. COHN: I guess we refer that to wordsmithing. If I can't read the sentence, it tells me there is something wrong with it. Would you give us leave to wordsmith this one a little bit?
When the SPLs become available, it is the intent of the NLM to update the repository drug information on a daily basis as a daily med. Additionally, as the full set of brand names is included in Rx Norm, some dispensed as-written prescriptions cannot be accommodated in e-prescribing.
There is also a new footnote which will have one change to it. It should read, the remaining 26 prescribed items that did not match to Rx Norm codes, not concepts, were primarily medical supplies that Rx Norm does not include, and drug delivery devices, e.g., steroid tapering dose packs and oral contraceptives which Rx Norm is just starting to include.
DR. HOUSTON: In the first change in that paragraph, the first 1,000 most common prescriptions written, it seems redundant. The first 1,000 --
DR. LUMPKIN: The 1,000 most prescribed. Scratch first.
DR. COHN: Thank you.
DR. MAYS: That is a different meaning. I thought when you said commonly, that what you meant is that there were 1,000, and in that 1,000, these were prescribed quite a bit, as opposed to 1,000 different ones.
DR. HOUSTON: But it is 1,000 different, because they said the NLM matched 974.
DR. COHN: Thank you. Now, recommendation 9.1 is unchanged, recommendation 9.2 now states, HHS should accelerate the promulgation of FDA's drug listing rule and hence the ability to support the correlation of NDC with Rx Norm, e.g., for passing daily updates of the SPL to NLM for inclusion of the daily med. The new sentence is, Timely rulemaking is critical to sustain the daily use of Rx Norm beyond the 2006 pilot tests I think we have to thank Russ for that addition to make a stronger statement there.
DR. LUMPKIN: What does it say about us that we can read through that sentence and understand what it says?
DR. COHN: That's right. There are a fair number of acronyms in there, if that is what you are wondering about us.
DR. LUMPKIN: But that's fine.
DR. COHN: Observations in ten are fine. Recommended action 10.1 now states, HHS supports NCPDP, HL-7 and others, especially including the prescriber community, in addressing SIG components in their standards. This should include preserving the ability to incorporate free text whenever necessary, e.g., for complex dosing instructions, and to address special cultural sensitivities, language and literacy requirements.
Observation 11 is unchanged, recommended action 11.1 has been modified. We will see this again in 12.1, where it says, HHS should insure that the NPI when it becomes available is incorporated as a primary identifier for dispensers in the NCPDP Script and other e-prescribing standards.
We see similarly in 12.1, where it also references -- this is for the prescriber, HHS should insure that the NPI when it becomes available is incorporated as a primary identifier for prescribers in the NCPDP Script and other e-prescribing standards.
Now, we are up to 13. We have major modification. I think we should just read through the whole observation, as well as the recommendations. NCVHS has observed that there were many impediments to full adoption and implementation of the HIPAA transactions and code set standards, with the requirement for pilot testing and e-prescribing, HHS has the opportunity to address any similar impediments that might arise with e-prescribing standard implementations.
As previously noted, while the foundation standards being recommended here are suitable for early industry adoption, there are several areas in the foundation standards that do not support all the MMA requirements. In addition, there are a number of patient and prescriber specific issues which must be studied and addressed, for example, acceptance and satisfaction. It must also be recognized that the success of the pilots will require much work by vendors to incorporate new standards and functionality into their applications, so that they can be pilot tested.
Okay with that?
DR. ROTHSTEIN: I have a question. I'm not sure what you mean in the last sentence, it must also be recognized. Are you saying that in order for the pilots to be successful, much work needs to be done? Or are you saying that if the pilots are successful, then much work needs to be done?
DR. LUMPKIN: The former.
DR. COHN: The former, though both are true.
DR. ROTHSTEIN: I think we could be clearer in making that statement, so that it is apparent that it is the former.
DR. LUMPKIN: Do you have a suggestion for that?
DR. ROTHSTEIN: We could say that in order for the pilots to be successful, much work will need to be done, blah, blah, blah.
DR. COHN: That sounds good.
DR. GREENBERG: It seems to me that what you are saying is that much work is going to have to be done by the vendors to incorporate new standards and functionality into their applications, so they can be pilot tested, right? It is a given, you can't pilot test them if they are not in the applications. So it seems redundant to say, in order for the pilots to be successful, we are saying we can't carry out these pilots without having this done.
DR. LUMPKIN: It may be that to really carry the intent -- because the comment that led to this was the fact that there is a lead time. We replaced the word so with before, so it reads, It must also be recognized that the success of the pilots will require much work by vendors to incorporate new standards and functionality into their applications before they can be pilot tested.
DR. VIGILANTE: Should we just say that we must be aware that they will require -- try to get this time thing in, that we must allow sufficient time for vendors to incorporate these standards into their applications so that they can be pilot tested.
DR. FITZMAURICE: But you can't start the pilot any later than will be required by law.
DR. VIGILANTE: I guess there is some urgency to get going, so that there is enough time to incorporate -- there is an issue around time.
DR. FITZMAURICE: So the vendors have to do that work, and so HHS should announce it early, so they have time to do that work.
DR. MAYS: What I had done was, at the end of the sentence, so they can be pilot tested in 2006. You are saying they are going to occur in 2006. You are letting them know they have to be ready for 2006.
DR. GREENBERG: I just got the picture that this is being addressed more to the Department than to the vendors. I think what needs to be said is that the vendors will need lead time to incorporate these new standards and functionality into their applications so that they can be pilot tested in 2006.
MS. FRIEDMAN: Just a quick comment to follow up on what Mike said, pilots go live January 1, 2006, so there is not a whole lot of lead time available.
DR. MAYS: What about saying that the Department should involve the vendors early so they will be ready on time? That is what you are really saying, talk to the vendors early so the vendors can be ready on time.
DR. LUMPKIN: We can maybe do that by saying, it must be recognized that success in the pilots will require much work by vendors to incorporate new standards and functionality in their applications before they can be pilot tested. Therefore, the Department should begin to work early with the vendors, or something to that effect.
DR. FITZMAURICE: But that will be a recommendation.
MR. BLAIR: Don't we already have a recommendation that comes very close to saying that?
DR. FITZMAURICE: We say widely publicized, but that is as close as we come.
DR. LUMPKIN: Maybe that should be 13 point something. We don't need to decide where it fits. Maybe between two and three would be the recommendation that HHS begin to work with the vendors as soon as possible to insure readiness for the initiation of the pilots in 2006.
MR. BLAIR: I support that.
MR. HUNT: In talking to Harry, he indicated that 2006 the pilots were supposed to start at the beginning of. I think it is important that in 2006 -- if I am a vendor, I'm thinking that means I've got until the end of 2006.
DR. LUMPKIN: HHS should begin to work with the vendors to insure initiation of the pilot tests by January 1, 2006.
DR. FITZMAURICE: It says a year's period, starting January 1.
DR. LUMPKIN: I think if we do that, we can just leave -- I think there is a difference between the observation and the recommendation.
MS. FRIEDMAN: Back up to the observation, I would like to take out sufficient for lead time. The lead time is what it is; it is set by statute.
DR. LUMPKIN: I think we can go back to the language that is there. It can just say, it must be recognized that the success of the pilots will require much work by vendors to incorporate new standards and functionality into their applications before they can be pilot tested. Then we go down to the new 12.3 with the recommendation.
DR. COHN: Are we okay with this at this point? John, thank you. 13.1 is unchanged, 13.2 is unchanged, 13.3., I think we just agreed that HHS should begin to work with the vendors to insure initiation of the pilot tests by January 1, 2006. It still isn't quite right.
DR. GREENBERG: To work with the vendors to insure incorporation of the standards into their products for pilot testing.
DR. LUMPKIN: How about to insure readiness?
DR. COHN: Yes, readiness for the pilot tests. That is probably a better general term. Do you want to say something like, HHS should begin to work immediately with the vendors? That is a slightly stronger word there. Are we comfortable with that?
The new 13.4, which used to be 13.3, HHS should identify and widely publicize specific goals, objectives, time lines and metrics to guide the design and assessment and increase industry awareness of the 2006 pilot test. I'd like to get rid of some of this stuff. HHS should include metrics that address economic, quality of care, patient safety and patient and prescriber satisfaction factors.
I am wondering, given all of our previous editions, whether we need the time lines and increased industry awareness in there, and whether that is necessary anymore, or have we already addressed that. I guess we dealt with the vendors, but we haven't for the rest of the industry. Maybe we need to keep it there.
DR. LUMPKIN: Let's leave it there.
DR. COHN: Okay. Are we okay with that one? Okay. The new 13.5. HHS should develop and widely disseminate information concerning the economic and quality of care benefits of e-prescribing, providing comprehensive education on implementation strategies, describe how e-prescribing fits into HIPAA privacy, and address other elements that contribute to successful and widespread prescriber adoption and patient acceptance.
I think we have a lot of questions about this one. Do you ant to start out?
DR. ROTHSTEIN: Yes. The main problem I have with this one is, the purpose of the pilot is to see if this works. My concern is that the way it reads now, we have already decided that issue. So I would propose a couple of changes. One is, before the word economic in the second line, I would add the word potential, so HHS should develop and widely disseminate information concerning the potential economic -- and then in the third line, describe how e-prescribing can be consistent with the privacy protections under HIPAA.
MR. BLAIR: Does that phrase already reference HIPAA?
DR. LUMPKIN: Yes, it does. It currently says, describe how e-prescribing fits into HIPAA privacy, then Mark's suggestion is fine.
DR. COHN: We don't have it quite fixed here.
DR. WARREN: I have a question about the intent. I thought the intent here was that after the pilots were done, they would disseminate the information. The way Mark has just edited it, it is before the pilot.
DR. COHN: I'm glad you are bringing this up. I looked at this one and it just seemed -- I couldn't figure this one out.
DR. WARREN: I think we need to decide, do we mean this before or after? The other Judy says, number four was before and number five was for after.
DR. COHN: Number five is making a conclusion that we haven't had the study done, so it gets a little awkward, since we don't know what the results of the pilots are going to show.
DR. LUMPKIN: The way we could do that is saying, HHS should develop and widely disseminate information concerning any economic and quality of care benefits of e-prescribing, provide comprehensive education and implementation strategy. There is language up there on how it is consistent with -- can be implemented consistent with, and address other elements that contribute to success and widespread prescriber adoption and patient acceptance.
DR. COHN: Okay. That hopefully is what the subcommittee meant. It is now. Implemented consistently or implemented consistent with?
DR. LUMPKIN: I think consistent with. Say all of this and then say at the end of that, Planning for this needs to begin immediately in order to -- success in the dissemination strategy requires planning prior to the completion of the pilot, prior to the end of the pilot, or something like that.
DR. VIGILANTE: Margaret, the second line in that 13.5 says, information concerning the any.
DR. LUMPKIN: Let me read it again. After the pilot test, HHS should develop and widely disseminate information concerning any economic and quality of care benefits of e-prescribing, provide comprehensive education on implementation strategy, describe how e-prescribing can be implemented consistent with the privacy protections under HIPAA, and address other elements that contribute to successful and widespread prescriber adoption and patient acceptance. Successful implementation requires planning for information dissemination to begin prior to the end of the pilot test.
DR. ROTHSTEIN: You want to plan your dissemination strategy as part of the design of the pilot test.
DR. LUMPKIN: Okay, begin prior to the initiation of the pilot test.
DR. STEINDEL: Is that a recommendation or an observation?
DR. LUMPKIN: It is a recommendation that they do it.
DR. COHN: I'm a little uncomfortable with this last sentence. I am reflecting back on Mark's comment, that we are assuming the results of the pilot before we do the pilot.
DR. STEINDEL: When you say this sentence, which sentence are you referring to?
DR. COHN: This is the very last sentence of recommendation 3.5. We are talking about work that is going to occur after the pilot tests, the dissemination plan.
DR. STEINDEL: You say it would begin before the end? The reason I feel uncomfortable with that is, I feel like it is starting to get too micro managing as to exactly when -- I think we could leave this up to HHS. They understand the intent from the rest of this.
DR. LUMPKIN: Okay. Can we live with what is there? Okay, Simon.
DR. COHN: One would observe at the bottom of page 13, there is a new footnote that relates to the question that had to do with commercial bias in observation 15. It now reads -- there was a question, where did it say in the regs about commercial bias. It is the conference agreement, where it says, report of the conference agreement on MMA notes that conferees intended for electronic prescribing to serve as a vehicle to reduce medical errors and improve efficiencies in the health care, but not to be used as a marketing platform or other mechanism to unduly influence the clinical decisions of physicians.
DR. LUMPKIN: That takes you to the end.
DR. COHN: I guess we are just going to have to agree on -- I think I would almost refer to John. At the very end it says, NCVHS offers to review and comment on plans for the 2006 pilot tests. I think that needs further wordsmithing.
MR. HUNT: Would be pleased to review?
DR. COHN: Okay.
MS. CARR: I would just suggest that an appendix be all these abbreviations.
DR. COHN: You may be right, but every time we have an abbreviation, we introduce it with the full text.
MS. CARR: I know that, but there are so many. I think one is going to want to be able to refer back.
DR. COHN: I think that is probably a good idea, so we can add that, glossary of terms.
DR. LUMPKIN: It has been moved by Simon, seconded by Jeff. All those in favor, signify by saying aye.
(Chorus of Ayes.)
DR. LUMPKIN: All those opposed? Any abstentions?
DR. COHN: Can I give a brief subcommittee report?
DR. LUMPKIN: Sure.
DR. COHN: This is going to be real quick. I want to thank you all for assisting us and supporting us through getting this done. I think the good news is that the subcommittee is not going to meet in September. We do have upcoming hearings on October 12 and 13. We will be talking about e-prescribing and we will be working with the committee offline to identify other issues that may need to be brought up at that point. We also have hearings previously identified for December 8 and 9, at which I'm sure there will be yet other issues.
Since our last meeting, the subcommittee has come up with additional dates for meetings. Right now we are currently planning on January 19 and 20. We are going to be polling the subcommittee members both for privacy and confidentiality, as well as standards and security, just to look at that week. We need to have another hearing on e-prescribing as it relates to privacy and confidentiality. I think there is some hope that we may be able to accommodate it all at once during that week. So we will be checking with people about re-looking at those dates to see if we can move things around a little bit to accommodate that.
DR. GREENBERG: Your first hearing on e-prescribing and privacy issues is November 19?
DR. COHN: Yes.
DR. GREENBERG: Then you are thinking of having another one in January?
DR. COHN: Yes, exactly. It will be sponsored by the Subcommittee on Privacy and Confidentiality.
DR. GREENBERG: Having to do with the association of --
DR. COHN: Exactly. We are going to try to do them all in the same week because of Mark's schedule. We also are holding for the subcommittee February 1 and 2, figuring that we do need to lead up to some March deliverables for everybody.
Other dates that we are currently holding are April 6 and 7, topic undetermined at this point, but likely not e-prescribing. Then we have some for July 6 and 7. We are likely to be polling some time in late May, early June again for other subcommittee dates.
So that is the report of the subcommittee. Again I want to thank the subcommittee for extraordinary work and service to the full committee in terms of getting this letter done. So thank you.
DR. LUMPKIN: Very good. Under Roberts Rules of Order, having completed the agenda, there is no need for a motion to adjourn. We just stand adjourned.
(Whereupon, the meeting adjourned at 1:50 p.m.)